Removal and Inactivation of Viruses by a Surface-Bonded Quaternary

Jan 24, 1990 - Downstream Processing and Bioseparation ... protein molecules on reducing the capacity and the rate of virus adsorption/inactivation pr...
1 downloads 0 Views 1MB Size
Chapter 13

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

Removal and Inactivation of Viruses by a Surface-Bonded Quaternary Ammonium Chloride I-Fu Tsao and Henry Y. Wang Department of Chemical Engineering, University of Michigan, Ann Arbor, MI 48109-2136

A quaternary ammonium chloride (QAC) covalently bound to alginate beads was shown to be capable of removing viruses from protein solutions. Bacteriophage T2 and herpes simplex virus type 1 (HSV-1) were used as model viruses in this investigation. The enveloped HSV-1 showed a higher susceptibility to surface-bonded QAC due to hydrophobic interaction. The elution experiments using 1% tryptone solution demonstrated that inactivation, instead of non-specific adsorption, is the main factor for virus titer reduction. Both equilibrium and kinetic studies were conducted in distilled water and protein (BSA, 0.5%) solution to observe the effect of protein molecules on reducing the capacity and the rate of virus adsorption/inactivation process. Heating thermolabile proteins at 60°C for 10 hours with various stabilizers has historically been used to inactivate any possible viral contaminants. However, the protecting effect of these stabilizers may minimize the destruction of the virus. Thermal inactivation and immobilized QAC inactivation procedures were compared usingβ-lactamase,a thermolabile enzyme, as a model protein. The activity ofβ-lactamasedropped down to 40% of the initial value after heating at 60°C for 10 hours using sucrose as a stabilizer. The virus titer diminished from 10 to 10 (PFU/ml) during the first four hours without further reduction. The titer of T2, a more hydrophilic virus, decreased only one order of magnitude and the recovery of β-lactamase activity was 70%. Computer simulation results demonstrating the effects of various process variables are also presented. 6

3

Quaternary a m m o n i u m chlorides ( Q A C ) are cationic surface-active agents w i t h antimicrobial activity (1). The major mode of action of Q A C was identified as the effects on cell permeability and cytolytic damage (2). K l e i n and Deforest (3) summarized the virucidal capacities of Zephiran ( a l k y l d i m e t h y l b e n z y l a m m o n i u m chloride) against v a r i o u s types of viruses. They reported that Zephiran can effectively inactivate l i p i d containing viruses, some non-lipid viruses and bacteriophages but is not effective against smaller but non-lipid viruses, such as picornaviruses. 0097-6156/90/0419-0250$06.00/0 © 1990 American Chemical Society In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

13. TSAO & WANG

Removal and Inactivation of Viruses

251

Quaternary ammonium chlorides display their antimicrobial activity even being i m m o b i l i z e d o n inert supports because they can act o n the membranes of various cells. The effects of surface-bonded organosilicon Q A C o n bacteria, yeasts, fungi, algaes have been extensively investigated (4, 5, 6, 7, 8). However, the activity against viruses of these i m m o b i l i z e d compounds has not been demonstrated. The presence of l o w level infectious viral contaminants has been one of the m a i n hurdles for wider application of genetically engineered therapeutic proteins such as blood-clotting factors, interferon, i n s u l i n , growth hormones, antibodies, and various clinically significant proteins such as tissue plasminogen activator. Heat treatment is a widely accepted method for sterilizing these complex protein solutions. It is well k n o w n that albumin can be pasteurized by heating at 60°C for 10 hours i n the presence of certain stabilizers (9). Unfortunately, similar attempts at other blood proteins (e.g. clotting factors) resulted i n marked reduction or e l i m i n a t i o n of their functional activities. S c h w i n n et a l . , (10) and Fernandes and Lundblad (11), modified the pasteurization procedure using h i g h concentrations (>30% w / v ) of polysaccharides as stabilizers. Unfortunately, the protein stabilizing effect of these polysaccharides likewise provides increased thermo-resistance to the viruses (12). Heat treatment of lyophilized clotting factor concentrates was also investigated (13). The phenomenon of v i r a l adsorption to various surfaces was extensively studied from an environmental standpoint as reviewed by Daniels (14) a n d Gerba (15) for prevention of various waterborne viral transmissions. The problem of virus removal f r o m complex protein solutions is very different f r o m that of sewage a n d d r i n k i n g water treatment processes because most protein molecules compete for the active sites of the adsorbents. Hence, both the adsorption rate a n d capacity diminish i n the presence of protein molecules (16). It is the intention of this paper to demonstrate and to compare the antiviral activity of a surfacebonded Q A C i n aqueous solutions against 2 m o d e l viruses w i t h and without the presence of proteins. The efficacy of the accepted antiviral thermo-inactivation was compared with the viral inactivation method by the surface-bonded Q A C treatment. Beta-lactamase was used as a thermolabile model protein (17), and bacteriophage T2 and herpes simplex virus type 1 (HSV-1, an enveloped animal virus) were used as model hydrophilic and hydrophobic viruses to test these chemical inactivation methods.

MATERIALS AND METHODS Chemicals, Three-(Trimethoxysilyl)-propyldimethyloctadecyl ammonium chloride (Si-QAC), known as D o w Corning 5700 Antimicrobial

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

252

DOWNSTREAM

PROCESSING A N D BIOSEPARATION

Agent, was provided by W . Curtis White (Dow C o r n i n g Corp., M i d l a n d , M i c h . ) . It is a methanolic solution containing 42 w t % of this active ingredient. Beta-lactamase was obtained from Sigma Chemical C o . (St. Louis, Mo.). Other chemicals were of reagent grade a n d were purchased from various commercial sources.

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

Organisms* E s c h e r i c h i a c o l i Β and bacteriophage T2 are regularly maintained i n our laboratories. BSC-1 cells and H S V - 1 strain 148 were obtained from D r . Charles Shipman, Jr. (Dental School, U . of M i c h i g a n , A n n Arbor, Mich.). The cultures are regularly passaged to maintain their viability. Preparation of Beads, Dried alginate/magnetite beads were prepared by a method similar to that described by Burns et al., (18). Barium chloride was used as a gel-inducing agent for better stability (19). The beads were further stabilized b y treating w i t h glutaraldehyde i n the presence of polyethyleneimine to avoid the dissolution problem (20). Beads w i t h diameters between 0.15 and 0.25 m m were obtained b y crushing and then systematically sieving the original spherical beads. Various concentrations of S i - Q A C solution were prepared by diluting the 42% active material i n distilled water at p H 5. After the beads were added to the S i - Q A C solution, the reaction temperature was raised to about 50°C for 10 minutes. Then, the p H was adjusted to 10.5 for another 10 minutes. The beads were then dried i n an oven (100°C), rinsed several times with sterile deionized water (pH 7.0) and stored at 4°C. Cell Culture» BSC-1 cells were g r o w n i n m i n i m a l essential m e d i u m ( M E M ) w i t h Earle salts supplemented with 10% fetal bovine serum (FBS) and 1.1 g/1 s o d i u m bicarbonate. Cells were passaged according to conventional procedures by using 0.05% trypsin plus 0.02 w t % ethylenediaminetetraacetic acid (EDTA) i n a HEPES-buffered balanced salt solution. Tissue culture flasks were incubated at 37°C i n a humidified 3% CO2 - 97% air atmosphere. Total cell counts were made using a Coulter counter equipped with a 100-μηι orifice and microscopic cell count. Titration of viruses* HSV-1 was assayed by using monolayer cultures of BSC-1 cells g r o w n i n 6-well multidishes. The cells were plated 3x10^ cells/well i n M E M ( E ) with 10% FBS and 1.1 g/1 sodium bicarbonate. After 24 hours, the cell sheet was about 75% confluent and was inoculated w i t h 0.2 m l of the virus suspension to be assayed and incubated for 1 hour to permit viral adsorption. The cell sheets were then overlaid w i t h 3 m l m e d i u m containing 0.5% methocel and incubated for another two days.

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

13. TSAO & WANG

253

Removal and Inactivation of Viruses

After aspiration of the overlay, the cells were stained w i t h crystal violet, and macroscopic plaques were enumerated. The assay procedures for T2 used here were described b y Rovozzo and Burke (21).

Assays. Samples collected i n all experiments were cooled and stored at

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

4°C, then the concentration of total protein i n the solution was assayed by the Bradford method (22), and the concentration of ß-lactamase was assayed according to Sykes and Nordstrom (23). Batch Experiments, D u r i n g these experiments, adsorbents and viruses were continuously mixed i n Erlenmeyer flasks by a gyratory shaker at 22°C. Reaction mixtures of k n o w n composition were made b y a d d i n g stock solution to 0.01 M T r i s / H C l buffer at p H 7.0. A l l stock chemical solutions were autoclaved and stored at 4°C. In the equilibrium studies, tests were conducted with various initial concentrations of adsorbents and viruses to determine the amount of virus adsorbed per u n i t g r a m of adsorbent a n d the v i r u s concentration remaining i n the solution at equilibrium. The time required to reach e q u i l i b r i u m was determined b y periodically s a m p l i n g over a 24-hour period. In the kinetic studies, samples were w i t h d r a w n at predetermined time intervals and assayed for virus titer.

RESULTS AND DISCUSSION Inactivation

of T2 phage

by QAC's

in Free

Solutions,

Susceptibility of bacteriophage T2 to Q A C is shown i n Table 1. Survivors could not be found i n solutions without the bovine serum albumin (BSA). These results demonstrated that bacteriophage T2 can be inactivated b y Q A C as w e l l as S i - Q A C solutions. The presence of protein molecules inhibited the activities of these antimicrobial agents. In fact, B S A was even coagulated i n the presence of high concentration (>0.05%) of S i - Q A C . Inactivation of Viruses by Surface-Bonded QAC, The attachment of this S i - Q A C to surfaces involves a rapid ion-exchange process w h i c h coats as a monolayer o n the bead surface. Then, the immobilization is further strengthened by the polymerization reactions (24). Table 2 shows the effects of dried alginate beads treated by various S i - Q A C concentrations. Zero percent means untreated beads and served as controls. When the titer was very l o w (4.7 χ 10 ), viruses were eliminated completely i n all cases including the control. This was due to non-specific adsorption. W h e n the titer was raised to 4.0 χ 10 , the adsorption capacities of treated beads were distinctly better than the control. For a titer as high as 2.0 χ 10 , it seems that the beads were nearly saturated with viruses i n all cases. 2

4

8

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

254

DOWNSTREAM

PROCESSING A N D BIOSEPARATION

Table 1. Antiviral Activity of Q A C Against T2 Phage

Disinfectant QAC*

D.W.*

7.4x105

QAC

0.5%BSA+

7.4 x l O

Si-QAC Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

Initial (PFU/ml)

Solution

Survivors (PFU/ml) 0 7.0 χ 10

5

D.W.

2.0x104

0

Si-QAC

D.W.

1.5x103

0

Control

D.W.

7.4 χ 10

7.1 χ 10

Λ

5

2

5

* 0.5% hexadecyltrimethyl ammonium chloride. 0.5% Dow Corning 5700 antimicrobial agent. # distilled water buffered by 0.01 M Tris/HCl, p H 7.0. + bovine serum albumin buffered by 0.01 M Tris/HCl, p H 7.0. No disinfectant was added to the control. Λ

Table 2. Effects of dried alginate beads treated by various concentrations of S i - Q A C Initial (PFU/ml)

10%

Survivors (PFU/ml) 1% 0.1% 0.01%

0% 0

4.7 x l O

2

0

0

0

4.0 x l O

4

0

0

2.0x10

2.0x108

1.5 x l O

7

1.3 χ 10

7

1.6 χ 10

0 2.0x10 7

1.3 χ 10

7

bead preparation: 2g of dried alginate beads in 20 ml of Si- Q A C solution, inactivation reaction: 2g of treated beads in 10 ml of 0.01 M Tris/HCl buffer solution, p H 7.0.

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

1.4 χ 10

4

2.8xl0

7

13. TSAO & WANG

255

Removal and Inactivation of Viruses

The titer reduction and adsorption capacities of the T2 phage and HSV-1 are compared i n Table 3. For similar initial titers (10 P F U / m l ) , the survivor titer of HSV-1 was at least 2 orders of magnitude lower than that of T2. F o r similar e q u i l i b r i u m titer remaining i n the solution (10 P F U / m l ) , the adsorption capacity (PFU/ml) of H S V - 1 was 2 orders of magnitude higher than that of T2. E v i d e n t l y , H S V - 1 is m u c h more susceptible to the surface-bonded Q A C than T2. Since H S V - 1 is an enveloped virus, the lipid bilayer surrounding the capsid binds strongly to the QAC-treated surface due to additional hydrophobic interaction. It should be noted that the adsorption experiments of T2 were carried out i n buffer solutions without proteins, while those of HSV-1 were i n buffered 1 v o l % FBS solution. Viruses can be considered as biocolloids w i t h surface charges that result from ionization of carboxyl and amino groups of proteins localized on the surface. A t a characteristic p H , defined as isoelectric point (pi), the virions exist i n a state of zero net charge. Isoelectric point of a virus may vary b y the type and the strain of the virus (25). The phage T2 (pl=4.2) possesses a net negative surface charge i n solutions of p H 7.0. O n the other hand, the Q A C treated bead renders a positively-charged surface. This suggests that electrostatic force m a y play an important role i n the adsorption process. However, the electrostatic force may not be the sole mechanism. Besides Brownian motion, the electrical double-layer (26), w h i c h is influenced by ionic strength and p H of the m e d i u m , m a y also facilitate the virus adsorption to the solid surface. Reduction of this double layer allows the van der Waals and hydrophobic to effect the adsorption of viruses to the immobilized Q A C surface. Quantification of these effects is generally difficult i n these complex protein solutions. 6

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

2

Elution Experiments, In addition to reversible adsorption, inactivation or degradation of viruses by various types of surfaces such as metal oxides (27), a l u m i n u m metal (28), magnetite (29), clays (30) a n d soils (31) have been reported. The mechanisms were identified to be either degradation of the capsid and/or the nucleic acid. However, such inactivation may be only specific to certain types of viruses. Bacteriophage T4 attached o n activated carbon can be reversibly eluted b y 1% tryptone solution (32). In this case, the majority of the adsorbed viruses could not be recovered by the tryptone elution (Table 4). The results suggest that the viruses were eluted off of the surfaces but i n an inactivated form (33). Adsorption Isotherms. Removal of T2 onto QAC-treated surfaces with and without the presence of BSA can be correlated using the Freundlich isotherms: q = KC

e

n

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

(1)

256

DOWNSTREAM

PROCESSING AND BIOSEPARATION

Table 3. Comparison of Titer Reduction and Adsorption

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

Capacity Between HSV-1 and T2 Phage Using Surface-Bonded Q A C

Virus

Initial (PFU/ml)

T2*

2.12 x l O

6

1.10 χ 10

4

1.50 χ 10

2

3.70 χ 10

5

2.03 x l O

6

6.67 χ 10

2

6.67 χ 10

2

2.03 χ 10

7

HSV-1

#

Survivors (PFU/ml)

Titer i n Solution Viruses Adsorbed (PFU/ml) (PFU/g)

* distilled water buffered by 0.01 M Tris/HCl, p H 7.0. # 1% FBS buffered by 0.01 M HEPES, p H 7.0.

Table 4. Elution of phage T2 after the adsorption/inactivation process using 1% tryptone solution

Initial (PFU/ml)

Survivors (PFU/ml)

After elution (PFU/ml)

2.1 χ 106

7.0 χ 103

1.1 x l O

4

99.5

1.8 χ 10

5

2.0 χ 10

2

3.9 χ 103

97.8

2.0 χ 10

4

4.2 χ 10

1

1.9 x l O

99.0

1.9x103

0

0

2

Inactivated T2/Total titer reduction (%)

100

inactivation reactions: 0.5 g of Si-QAC treated beads in 5 ml 0.01 M Tris/HCl buffer solution, p H 7.0.

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

13. TSAO & WANG

Removal and Inactivation of Viruses

257

where q is the amount of viruses removed and C is the virus titer i n equilibrium remaining i n the solution, Κ and η are coefficients w h i c h can be determined by linear regression. Typical isotherms for removal of viruses by QAC-treated beads are shown i n Figure 1. The value of η is close to one i n both cases. A significant reduction i n the adsorption capacity is observed i n 0.5% B S A solution because the B S A molecules interfere with the adsorption of the viruses. In Figure 2, kinetics of T2 removal using Q A C treated beads is presented. It is obvious that the competitive adsorption between viruses and BSA molecules also reduced the adsorption rate. In both cases, viruses were inactivated rapidly at the initial 2 hour mark and titer reduction s l o w e d d o w n after that. This inconsistency w i t h the first-order inactivation model may be due to various interfering mechanisms such as d i s p l a c e m e n t , m o l e c u l a r o r i e n t a t i o n , m u l t i l a y e r effects, surface heterogeneity, and virion clumping.

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

e

Adsorption/lnactivation of T2 Phage in a ß-Lactamase Solution. The denaturation or unfolding of a protein leads to loss of its functional activity. The activation energy of the protein unfolding process can be increased i n the presence of sucrose (34). The activity of ß-lactamase, a model protein, dropped d o w n to 40% of the initial value after heating the mixture at 60°C for 10 hours using sucrose (0.8 g/ml) as a stabilizing agent (Figure 3). The total amount of soluble proteins decreased because of coagulation. The decline of ß-lactamase activity agreed w i t h that of the total protein. These experimental results compared favorably with various sucrose stabilization studies of thermolabile proteins using blood clotting factors (10, 11). It was assumed i n those studies that the treatment can render the protein solutions free of hepatitis infection. However, Figure 3 shows the T2 titer also diminished from 10*> to 10^ d u r i n g the first four hours without further reduction. Inactivation of phage T2 i n ß-lactamase solution b y Q A C - t r e a t e d beads is shown i n Figure 4. The initial T2 titer was 3.0 χ 1 0 P F U / m l . A quantity of 0.8 g of beads were mixed with 10 m l of ß-lactamase solution. Fifteen percent of the viruses survived this treatment. The amount of total protein i n the solution was 80% of the initial value after the adsorption process, while the recovery of ß-lactamase activity was at least 70%. It was the purpose of this experiment to demonstrate that Q A C treated beads can effectively remove viruses from a protein solution without significantly losing the activity of the protein. Optimal adsorption condition and mode of operation ought to be determined by studying the interactive effects of p H , ionic strength, and temperature of the solution, with the specific types of virus and protein of interest. 6

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

258

DOWNSTREAM

10° J

1

10°

10

1

10

1

1

10

2

PROCESSING AND BIOSEPARATION

1 3

10

1 4

10

1

10

5

1 6

Virus Remaining (PFU/ml)

Figure 1. Equilibrium isotherms of phage T2 inactivation using surface-bonded Q A C 100

o

A

0

1

2

3 Time

4

(hr.)

* initial T2 titer : 5.5 χ 1 0 PFU/ml 6

Figure 2. Kinetic study of phage T2 removal using surface-bonded Q A C

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

10

7

13. TSAO & WANG

100

259

Removal and Inactivation of Viruses

initial T2 titer : 5.0 χ 10 PFU/ml total protein : 0.5 mg/ml ß-lactamase: 885 IU/ml 6

10

5

ι* ID

LL ç

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

!io

4

"c cd

ε

Φ ce



> Il03

Figure 3. Thermal inactivation of phage T2 i n ß-lactamase solution using sucrose as stabilizer

total protein

ß-lactamase

initial T2 titer : 3.1 χ 10 PFU/ml total protein : 0.5 mg/ml ß-lactamase: 885 IU/ml 6

phage T2

Time (hr.)

Figure 4. Removal of phage T2 from ß-lactamase solution using surface-bonded Q A C

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

260

DOWNSTREAM

PROCESSING AND BIOSEPARATION

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

Simulation Studies of Virus Removal Usina Adsorption Column. A fixed-bed adsorption has several advantages over batch and continuous stirred tank reactor (CSTR) because the rates of adsorption depend on the concentration of viruses i n solution. This point is especially important for virus removal because of the l o w concentration of viral contaminants. The design of a fixed-bed adsorption column involves estimation of the shape of the breakthrough curve and the appearance of the breakpoint. C o m p u t e r s i m u l a t i o n studies were done here to demonstrate the performance of a virus adsorber using the surface-bonded Q A C beads which have a higher binding affinity for viruses over other proteins. The diffusion model of the system can be described mathematically by sets of material balance equations together with appropriate boundary and initial conditions: Equation of Continuity (2) Adsorption Rate RAi - ^ f - ( C i - C e i )

( 3 )

Adsorption Equilibrium

qi = Kf C i

(4)

Initial Conditions

Q(z,0) = 0

(5)

qi(z,0) = 0

(6)

Q(0,t) = Q o

(7)

e

Boundary Conditions

3Ci(L,t)

n

= 0

(8)

3 z

where C i , C i , and Q are the concentrations of i adsorbate i n the bulk solution, at the interface, and of the influent, ν is the linear velocity, L is the bed length. Linear adsorption isotherms (n=l) are assumed for both virus and total protein. The equilibrium constants Κ were obtained from batch experiments. It was also assumed that the complex proteins can be considered as a single component, no radial concentration gradient, and diffusion coefficients, fluid viscosity and density remained constant. e

0

t

h

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

13. TSAO & WANG

Removal and Inactivation of Viruses

261

The quantity D L is the longitudinal dispersion coefficient of viruses and can be determined b y the empirical correlation given b y C h u n g and W e n (35) as a function of Reynolds number (Re), density and viscosity of the fluid. P?k = μ

Re 0.2 + 0.11 Re0.48

( 9 )

The mass transfer coefficient k is estimated b y the correlation of dimensionless j , or Colburn factor, with Sherwood number (Sh), Schmidt number (Sc), and v o i d fraction as described by Cookson (36). Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

c

j=

Sh

=

Re Sc /

J

1

kç (V)2/3

3

v

D

(10)

and j = Be Re- / 2

3

(11)

These equations were solved numerically using finite difference method with double precision. Figures 5(a) and 5(b) show the simulated breakthrough curves of both total protein and H S V - 1 respectively. It should be noticed that the dimensionless time scales i n these t w o figures differ b y four orders of magnitude. The breakpoint of HSV-1 is the operating endpoint at which the effluent from the adsorption column can no longer meet the desired sterilization criterion. Since the HSV-1 has a much higher affinity to the bead surface, the breakpoint of HSV-1 appears much later than that of the total protein. To optimize the protein recovery, one should improve the design of the bead surface (better selectivity, higher loading capacity), size, and operating parameters of the filter to further delay the breakpoint of the virus elution. A stochastic approach to model the removal process may be more appropriate i n l o w concentrations of viruses. The effects of desired sterilization criterion o n total protein recovery and the amount of adsorbent required are demonstrated i n Figure 6. Stringent sterilization criterion (10 ) can only be achieved w i t h reduced protein recovery based on our current design of the beads. 1

-3

CONCLUSIONS The surface-bonded Q A C can effectively adsorb a n d inactivate viruses based o n our initial experimental results. H S V - 1 , an enveloped virus, is 1

The physical parameters used for simulation are listed in Table 5.

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

262

DOWNSTREAM

PROCESSING

AND BIOSEPARATION

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

Table 5. Physical Parameters Used for Simulation Studies

Adsorbates parameters: HSV-1

FBS

Equilibrium constant K ( m l solution/ml adsorbent)

5.0 χ 10

Diffusion coefficient D(cm /sec.)

8.0 χ 10'

Influent titer (PFU/ml) or concentration (mg/ml)

1.0 χ 10

5.25

4

8

6.0 χ 10"

2

5

C o l u m n parameters: column diameter (cm)

: 3

column porosity (-)

: 0.5

bead diameter (cm)

: 0.02

bead density (g/ml)

: 2.11

Fluid parameters: viscosity (centipoise)

: 1.20

density (g/ml)

: 1.01

flow rate (ml/min)

: 1.0

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

500

7

13.

TSAO & WANG

263

Removal and Inactivation of Viruses

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

(a) BREAKTHROUGH CURVE OF TOTAL PROTEIN

Ο

0

2

4

6

8

10

DIMENSIONLESS TIME (T) ~

(b) BREAKTHROUGH CURVE OF HSV-1

DIMENSIONLESS TIME (T)

Figure 5. Simulated adsorption breakthrough curves of total protein and HSV-1

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

264

DOWNSTREAM

PROCESSING

AND

BIOSEPARATION

Sterilization criterion (PFU/ml)

Figure 6. The effect of sterilization criteria on the protein recovery and the required amount of adsorbent more susceptible than the non-enveloped bacteriophage T2 to the Q A C treatment. However, as a non-specific adsorption process, both the rate and capacity were reduced due to the competitive b i n d i n g of the protein molecules. Thermo-inactivation and surface-bonded Q A C treatment were compared i n terms of titer reduction and remaining functional activity of a model protein, ß-lactamase. Process modeling and computer simulation enable us to predict the breakthrough curves of a virus adsorption column. Choosing a specific sterilization criterion has to be compromised w i t h reduced protein recovery if adsorption has to be used for the removal of viruses i n protein solutions.

ACKNOWLEDGMENT The partial financial support of the N a t i o n a l Science Foundation is acknowledged.

Notation Be C Cq D E>L Dp j

= = = = = = =

constant depending on void fraction. fluid phase-concentration, virions/cm or μg/cm . fluid phase inlet concentration, virions/cnv* or μg/cm . diffusion coefficient, cm^/sec. longitudinal dispersion coefficient, cm /sec. mean diameter of adsorbent, cm. dimensionless Colburn factor. 3

3

3

2

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

13. TSAO & WANG

Removal and Inactivation of Viruses

Κ kc L η pe q R Re

= = = = = = = =

volume equilibrium constant, cmfycnA mass transfer coefficient, cm/s. length of column, cm. parameter in Freundlich isotherm. VL/E>L, Peclet number. solid phage concentration, virions/cm^ or μg/cnA mean radius of adsorbent, cm. vDp/υ, Reynolds number.

RA Sc Sh Τ t U ν

= = = = = = =

adsorption rate, virions/cnvVsec or Mg/cnvVsec. υ/D, Schmidt number. kcDp/D, Sherwood number. v t / L , dimensionless time. time, sec. C / C , dimensionless fluid-phase concentration, average linear velocity, cm/sec.

265

0

Greek Letters ε

=

void fraction, cnvVcm^

p

=

fluid bulk density, g/cnv*

μ

=

absolute viscosity, g/cm/sec

υ

=

kinematic viscosity, cm /sec 2

LITERATURE CITED 1. 2. 3. 4. 5. 6. 7. 8. 9.

Petrocci, A. N. Surface-active agents: quaternary ammonium compounds, In S.S. Block (ed.), Disinfection, Sterilization and Preservation, 3rd ed. Lea and Febiger, Philadelphia, PA. 1983. p. 309-334. Hugo, W. B. "The mode of action of antimicrobial agents" J. Appl. Bacteriol. 1967. 30: 27-60. Klein, M. and Deforest, A. Principles of viral inactivation, In S.S. Block (ed.), Disinfection, Sterilization and Preservation, 3rd ed. Lea and Febiger, Philadelphia, PA. 1983. p. 422-434. Walters, P. Α., Abbott, Ε. Α., and Isquith, A. J. "Algicidal activity of a surfacebonded organosilicon quaternary ammonium chloride" Appl. Microbiol. 1972. 25: 253-256. Isquith, A. J., Abbott, E. A. and Waters, P. A. "Surface-bonded antimicrobial activity of an organosilicon quaternary ammonium chloride" Appl. Environ. Microbiol. 1972. 24: 859-863. Isquith, A. J. and McCollum, C. J. "Surface kinetic test method for determining rate of kill by an antimicrobial solid" Appl. Environ. Microbiol. 1978. 36: 700-704. Speier, J. L. and Malek, J. R. "Destruction of microorganisms by contact with solid surfaces" J. Colloid Interface Sci. 1981. 89: 68-76. Nakagawa, Y., Hayashi, H., Tawaratani, T., Kourai, H., Horie, T. and Shibasaki, I. "Disinfection of water with quaternary ammonium salts insolubilized on a porous glass surface" Appl. Environ. Microbiol. 1983. 47: 513-518. Gellis, S. S.; Neefe, J. R.; Stokes, J.; Stong, L. E.; Janeway, C. A. and Scatchard, G. "Chemical, clinical, and immunological studies on the virus of homologous serum hepatitis in solutions of normal human serum albumin by means of heat" J. Clin. Invest. 1948. 27: 239-244.

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

266 10. 11. 12. 13. 14.

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

15. 16. 17. 18. 19.

20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33.

DOWNSTREAM PROCESSING AND BIOSEPARATION Schwinn, H., Heimburger, N., Kumpe, G. and Herchenhan, B., "Blood coagulation factors and process for their manufacture" U.S. Patent, 4,297,344. 1981. Fernandes, P., Lundblad, J. L. "Pasteurized therapeutically active protein composition" U.S. patent, 4,440,679. 1984. Ng, P. K. and Dobkin, M. B., "Pasteurization of antihemophilic factor and model virus inactivation studies" Thromb. Res. 1985. 39: 439-447. Rubinstein, Α., "Heat treatment of lyophilized blood clotting factor VIII concentrate" U.S. patent, 4,456,590. 1984. Daniels, S. L. "Mechanisms involved in sorption of microorganisms to surfaces" In G. Bitton and N.C. Marshall (ed.), Adsorption of Microorganisms to Surfaces. John Wiley and Sons, Inc., New York. 1980. p. 7-58. Gerba, C. P. "Applied and theoretical aspects of virus adsorption to surfaces" Adv. Appl. Microbiol. 1984. 30: 133-168. Lipson, S. M. and Stotzky, G. "Effect of proteins on reovirus adsorption to clay minerals" Appl. Environ. Microbiol. 1984. 48: 525-530. Smith, J. T. "R-factor gene expression in gram-negative bacteria" J. Gen. Microbiol. 1969. 55:109-120. Burns, M., Kvesitadze, G. I., and Graves, D., "Dried alginate/magnetite spheres: a new support for chromatographic separations and enzyme immobilization" Biotech. Bioeng. 1985. 27: 137-145. Paul, F. and Vignais, P. M. "Photophosphorylation in bacterial chromatophores entrapped in alginate gel: improvement of the physical and biochemical properties of gel beads with barium chloride as gel-inducing agent" Enzyme Microb. Technol. 1980. 2: 281-287. Birnbaum, S., Pendleton, R., Larsson, P-O., Mosbach, K., "Covalent stabilization of alginate gel for the entrapment of living whole cells" Biotech. Lett. 1982. 3: 393-400. Rovozzo, G. C. and Burke, C. N. "A Manual of Basic Virological Techniques" Prentice-Hall Inc., Englewood Cliffs, New Jersey. 1973. Bradford, M. "A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding" Anal. Biochem. 1985. 72: 248-254. Sykes, R. B. and Nordstrom, K. "Microiodometric determination of β-lactamase activity" Antimicrob. Agents Chemotheraphy. 1972. 1: 94-99. Malek, J. R. and Speier, J. L. "Development of an organosilicon antimicrobial agent for the treatment of surfaces" J. Coated Fabrics. 1982. 12: 38-45. Zerda, K. S. Ph.D. Dissertation. Baylor College of Medicine, Houston, Texas. 1982. Verwey, E. J. W. and Overbeck, J. G. "Theory of the stability of lyophobic colloids" Elsevier, Amsterdam. 1984. Murray, J. P. and Laband, S. J. "Degradation of poliovirus by adsorption on inorganic surfaces" Appl. Environ. Microbiol. 1979. 37: 480-486. Murray, J. P. "Physical chemistry of virus adsorption and degradation on inorganic surfaces" U.S. Environmental Protection Agency, Cincinnati, Ohio. 1980. Atherton, J. G., Bell, S. S., "Adsorption of viruses on magnetic particles" Water Res. 1983. 17: 943-953. Tayler, D. H., Bellamy, A. R. and Wilson, A. T. "Inactivation of bacteriophage R17 and reovirus type III with the clay mineral allophane" Water Res. 1980. 14: 339346. Yeager, J. G. and O'Brien, R. T. "Structural changes associated with poliovirus inactivation" Appl. Environ. Microbiol. 1979. 38: 702-709. Cookson, J. T., North, W. J. "Adsorption of viruses on activated carbon" Environ. Sci. Technol. 1967. 1: 46-52. Tsao, I-F., Wang, H. Y., Shipman, Jr., C. "Interaction of infectious viral particles with a quaternary ammonium chloride (QAC) surface" Biotech. Bioeng. 1989. 34, 5: 639-646.

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

13. TSAO & WANG 34. 35. 36.

Removal and Inactivation of Viruses

267

Lee, J. C. and Timasheff, S. N. "The stabilization of proteins by sucrose" J. Biol. Chem. 1981. 256: 7193-7201. Chung, S. F. and Wen, C. Y. "Longitudinal dispersion of liquid flowing through fixed and fluidized bed" AIChE J. 1968. 14: 857-866. Cookson, J. T. "Removal of submicron particles in packed beds" Environ. Sci. Technol. 1970. 4: 128-134.

Downloaded by UNIV OF ARIZONA on December 14, 2012 | http://pubs.acs.org Publication Date: January 24, 1990 | doi: 10.1021/bk-1990-0419.ch013

RECEIVED October 27, 1989

In Downstream Processing and Bioseparation; Hamel, J., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1990.